BLE long range (coded phy) is same than 1MBit

[RichardHdrd]

Dear Community,

I implemented the beacon-example with S112 and 1MBit mode and with S140 and coded-phy 125k sending one beacon per second continuously at +0dBm (sd_ble_gap_tx_power_set).

I implemented a BLE scanner with S140 and choice of 1MBit and coded-phy 125k mode reception, the latter using the cfg:

static ble_gap_scan_params_t const m_scan_param_coded_phy =
{
    .extended       = 1,
    .active        = 0x00,
    .interval      = NRF_BLE_SCAN_SCAN_INTERVAL,
    .window        = NRF_BLE_SCAN_SCAN_WINDOW,
    .timeout       = NRF_BLE_SCAN_SCAN_DURATION, // No timeout.
    .scan_phys     = BLE_GAP_PHY_CODED,
    .filter_policy = BLE_GAP_SCAN_FP_ACCEPT_ALL,
};

Now performing tests, I get similar results for both, 1MBit and coded-phy 125k mode.

As I can well understand, that the RSSI values don't differ as there is no difference in expected transmission/reception signal strength, I do not understand, why devices in both modes stop detecting beacons at an RSSI value of about -92dBm.

The vendor (FANSTEL) states for the BM833(A) module which uses a nordic 52811 chip:

BLE Sensitivity is increased to -97 dBm at 1Mbps and -104 dBm at 125 kbps with CODED PHY.

So is there any reason, why reception stops at RSSI -92dBm, especially for coded-phy mode?

Best regards,

Richard

[RichardHdrd]

Hi Sigurd,

I'm using SDK version 17.0.2 and Softdevices Version 7.2.0

I tested in a line-of-sight between the devices, but as our main use-case will be indoors, we did indoor testing.

As for sending more frequently and setting .active to 0x01, I will try and report back, but if sending beacons more frequently would solve the issue, we'd run into the problem, that our sender is battery driven and sending beacons more often reduces battery life-time...

We currently do not need any connection as we provide no service (we just transmit some values within the beacon), so we test for scanning/advertising only.

Best regards,

Richard

[Sigurd]

RichardHdrd said:

I'm using SDK version 17.0.2 and Softdevices Version 7.2.0

 Good. That is the latest version.

RichardHdrd said:

I will try and report back,

 Ok. What HW do you have only BM833(A) modules ? or do you have  e.g. a nRF52840DK as well?

[RichardHdrd]

I have a nRF52833DK, a FANSTEL EV-BM833, a nRF52840DK and a FANSTEL EV-BM833A

These, I can used as Centrals and they produce similar test-results.

I use a BM833A as Peripheral.

I tried as you suggested (send the bacon packet more frequently / try to change .active to 0x01)

But still, I have the same issues (reception stopping at RSSI about -92dBm also for coded-phy BLE and the achievable range is about 30m in the open with line of sight, which is factors away from the observed 1km plus.

There are some significant issues, compared to the link with the BLE-range test provided by you.

Quote:

With 1M PHY, the scanner logs show that we can get adv packets with a signal strength of around -93 dBm. With Coded PHY, the scanner can still detect adv packets of signal strength around -101 dBm.

So, the 1MBit mode drops out at similar RSSI value (-92dBm here, -93dBm measured). But, when I use the 1MBit non-extended mode, i measure only about 30m reception range for adv packets, then I hit -92dBm and that's it.

BTW: When measuring at 10cm (at 0dBm sender amplification) I have an RSSI of -37dBm, at 1m the value is -52dBm.

Finaly, the long-range coded-phy mode seems to behave identically to the 1MBit non-extended mode.

Best regards,

Richard

[Sigurd]

Hi,

Your results seem to be expected, as the RSSI will saturate around 90 dbm, and RSSI values outside the range listed in ths PS are not valid.

nRF52811:

https://infocenter.nordicsemi.com/index.jsp?topic=%2Fps_nrf52811%2Fradio.html&anchor=unique_356012096

[RichardHdrd]

Hi,

So, when my RSSI measures are OK, I still wonder, why I get only about 30m range (40m when measuring outside) for BLE coded-phy. Is there any visible reason for that behaviour; how important is a clear line of sight when dealing with bluetooth?

Best regards,

Richard

[Sigurd]

RichardHdrd said:

how important is a clear line of sight when dealing with bluetooth?

 It depends on the environment, but if you are trying to get 100 meters+ with range, then line of sight is pretty important.

From this page, "Path loss, or path attenuation, occurs naturally over distance and is impacted by the environment in which the signal is being transmitted. Obstacles between the transmitter and the receiver can deteriorate the signal. Attenuators can be anything from humidity and precipitation, to walls, windows, and other obstacles made of glass, wood, metal, or concrete, including metal towers or panels that reflect and scatter radio waves. While radio waves can pass through objects, the amount of attenuation and effective path loss varies with the type and density of the obstruction."